Breadcrumb: Home > Applications > Lab Research
Updated:2026-11-02
Views: A filler can lift the shielding effectiveness of an electromagnetic interference film to 26 dB and still fail in service: the film delaminates on repeated bending and loses attenuation after an acid or alkaline soak. The usual cause is that fiber and filler are held together by physical adsorption alone, with no chemical bond across the interface. A 2026 study from Wuhan Textile University in Surfaces and Interfaces addresses this with UV crosslinking, and its methods section states the equipment, wavelength, power, time and cumulative energy in enough detail to repeat. This article breaks down the four-step process, reads the three key parameters, and lists the points where reproduction usually goes wrong.
Groups working on EMI shielding films have been asking the same question for the past two years: the filler is in, the shielding effectiveness is up by 26 dB, so why does the film fail on bending and decay after an acid or alkaline soak? The problem is usually not the formulation but the interface: fiber and filler are held together by physical adsorption alone, with no chemical bonds. The work by Guo Jianing, Dong Xinrui and colleagues at the State Key Laboratory of New Textile Materials and Advanced Processing Technology, Wuhan Textile University, published in Surfaces and Interfaces in 2026, answers with UV crosslinking - and in the methods section they state the equipment, wavelength, power, time and cumulative energy in full, identifying the instrument as a LUYOR UCL-3500.
The paper deserves a separate breakdown for two reasons. Unlike most of the literature collected on this site, the crosslinking step is reported as a complete five-tuple that can be reproduced directly (model, wavelength, power, time, cumulative energy), not a vague "UV irradiation for several minutes". And the performance path is clean: the same matrix, the same batch of filler, one extra 30-minute UV step, and tensile strength, shielding effectiveness and environmental stability all move up together, so the causal chain is visible.
| Item | Setting in the paper |
|---|---|
| Paper | Guo J. et al., Surfaces and Interfaces 100 (2026) 110979, DOI 10.1016/j.surfin.2026.110979; online 10 October 2026 |
| Matrix | PVA-co-PE nanofibers (100-200 nm diameter, made by melt blending and phase separation) plus UHMWPE fiber (FT-103) interpenetrating network, feed ratio 1:10 |
| Functional filler | Fe3O4@RGO (hydrothermal synthesis, Fe3O4 anchored on the RGO surface), mass ratio to PVA-co-PE 3:97 |
| Film forming | Vacuum filtration of a multi-component suspension |
| Crosslinking system | TMPTMA 1.0% (V/V, trimethylolpropane trimethacrylate) plus DCP 0.2% (V/V, dicumyl peroxide), ethanol solution |
| Crosslinking equipment | LUYOR UCL-3500 UV crosslinker |
| Irradiation parameters | Wavelength 365 nm; power setting 6500 μW/cm²; time 30 min; cumulative energy 4.68×105 J/m² |
| Post-treatment | Air dried in a fume hood to constant mass; mass essentially unchanged before and after crosslinking |
| Key results | Tensile strength 13.3 MPa; EMI SE 37.1 dB; SSE 1515.53 dB·cm⁻¹; SSE/t 6314.68 dB·cm²·g⁻¹ |
One correspondence is easy to miss: the paper writes "UCL-3500, wavelength 365 nm". In the LUYOR line the same instrument is built in three wavelength versions, distinguished by tube - UCL-3500 (254 nm), UCL-3500M (302 nm) and UCL-3500L (365 nm). The unit actually used in the paper is therefore the 365 nm UCL-3500L, not the default 254 nm version. When ordering or reproducing, match on the wavelength version, not on the model number alone.
The introduction puts the problem plainly. Between an insulating polymer fiber and a conductive filler, most published work relies on physical adsorption, and that bond is weak: under repeated bending, stretching or aggressive environments the filler detaches and performance decays. Pushing the filler loading higher to reach high shielding effectiveness brings its own penalty - a film that behaves more like an inorganic body, with worse mechanics and processability. These two are exactly the practical pain points of flexible shielding films.
UV crosslinking fits because it works on two levels at once. It welds fiber to fiber and fiber to filler, upgrading the interfacial bond from van der Waals forces to covalent bonds; and the crosslinked network compacts and tightens the conductive filler, making the conductive paths more continuous. One step benefits both the interface and the conductive network - which is why the paper reaches 37.1 dB at a comparatively low filler loading.
| Step | Key operation | Where it goes wrong |
|---|---|---|
| 1. Dispersion and filtration | PVA-co-PE nanofibers dispersed in water/isopropanol (mass ratio 3:50:50), beaten at high speed to 3.0 wt% fiber content; Fe3O4@RGO mixed in at 3:97; UHMWPE fiber added at 1:10; vacuum-assisted filtration into a film | Filtering before the nanofibers are opened up leaves agglomerates in the film, which become stress concentrators after crosslinking |
| 2. Soak in the crosslinking bath | Wet film immersed in ethanol solution containing TMPTMA 1.0% (V/V) and DCP 0.2% (V/V) | Raising the crosslinker concentration at will does not give a stronger bond; it leaves residue and a tacky surface |
| 3. UV crosslinking | Film taken out and placed directly in the UCL-3500, 365 nm, 6500 μW/cm², 30 min | Going into the chamber wet, not turning the film over, controlling by time rather than by energy |
| 4. Drying | Air dried in a fume hood to constant mass; the paper reports mass essentially unchanged, so the crosslinker has reacted or evaporated with no obvious residue | Oven drying at high temperature destroys the crosslinked network already formed |
Note that there is no drying step between steps 2 and 3 - the film goes into the chamber still wet with the ethanol solution carrying the crosslinker. This matters for reproduction: in the wet state radicals live longer and oxygen diffuses into the film less readily, so oxygen inhibition is suppressed substantially. Converting the process to drying before irradiation typically lowers the reaction efficiency and requires the dose to be established again.
The initiator here is DCP and the crosslinker is TMPTMA; both are organic molecules whose absorption sits on the long-wave UV side. 365 nm also penetrates a polymer film far more deeply than 254 nm (at 254 nm most organic films absorb the light within tens of micrometers, which tends to produce a layered structure with a crosslinked skin and an untreated interior). For filtered films from tens to a hundred micrometers thick, the longer wavelength is what carries the reaction through. This is also why the 254 nm version used by default for nucleic acid membrane crosslinking in biology labs is not suitable for copying here.
6500 μW/cm² is 6.5 mW/cm². It is the intensity level inside the chamber, and it shifts with tube aging, the number of tubes and the position of the sample. What belongs in a methods section, and what another group needs in order to reproduce the work, is the cumulative energy.
At 6.5 mW/cm² for 30 min (1800 s), the cumulative energy works out to 6.5 mW/cm² × 1800 s = 11 700 mJ/cm² = 1.17×105 J/m². The paper reports 4.68×105 J/m², exactly four times the calculated value. This usually comes from one of two conventions: the power setting is not the same quantity as the energy accumulated at several points inside the chamber (multi-tube models often sum over several detectors), or a conversion is being made between the power setting and the measured irradiance at the plane.
So do not copy "6.5 mW/cm² × 30 min" mechanically. Work from the instrument's cumulative energy reading, or measure the intensity at the film plane with a UV radiometer and integrate it yourself. If the model on hand has no energy mode, measure the actual intensity at the film plane in mW/cm², divide the target dose by that intensity to get the time, and recalibrate after every tube change.
The mechanism the paper sets out is a closed loop: DCP decomposes under UV to give radicals, which abstract hydrogen from PVA-co-PE and UHMWPE to form macromolecular radicals; the three double bonds of TMPTMA couple those radicals into stable C-C crosslinks; both matrices also self-crosslink; and the oxygen-containing groups on the RGO surface hydrogen-bond with the hydroxyls of PVA-co-PE. Three paths overlap, and the interfacial force is what finally rises.
| Characterization | Before crosslinking | After crosslinking | Note |
|---|---|---|---|
| FTIR C=C peak (1634 cm⁻¹) | relatively strong | weaker and shifted | the TMPTMA double bonds took part in the reaction |
| FTIR C=O peak (1732 cm⁻¹) | relatively strong | weaker and shifted | as above, supporting evidence that crosslinking occurred |
| FTIR O-H peak | 3431 cm⁻¹ | shifted to 3267 cm⁻¹ | hydroxyls move from the free to the associated state as hydrogen bonds form |
| XPS C-C binding energy | 283.9 eV | 284.9 eV | new C-C bonds form between PVA-co-PE and UHMWPE |
| XPS O-H binding energy | 530.8 eV | 531.7 eV | new O-H appears at the surface, consistent with FTIR |
| TG curve | two decomposition stages | onset temperature and peak decomposition rate temperature both rise | the crosslinked network delays thermal decomposition of PVA-co-PE and UHMWPE |
Without XPS on hand, tracking the intensity of the 1634 cm⁻¹ and 1732 cm⁻¹ peaks in FTIR is a low-cost criterion for crosslinking; add a gel content (or swelling ratio) measurement and different doses can be compared quantitatively, without running the full characterization set every time.
| Sample | Tensile strength | Relative change |
|---|---|---|
| PVA/UPE matrix (no filler) | 5.1 MPa | - |
| PVA/UPE/FeO/RG (filler, not crosslinked) | 7.9 MPa | +54.9%, from hydrogen bonding between components |
| PVA/UPE/FeO/RG-X (crosslinked) | 13.3 MPa | a further ~68% over the uncrosslinked film (stated in the paper as more than 1.6 times) |
Elongation at break falls clearly after crosslinking: crosslink points resist sliding, chain motion is restricted, and the stress needed for a given strain rises. The paper notes, however, that the elongation of the crosslinked film is still close to that of the filler-free PVA/UPE base film - flexibility is not lost, it simply returns from the extra ductility contributed by the filler to the level of the matrix itself. Modulus rises markedly. In compression, 100 cycles were run at each of 20%, 40%, 60% and 80% strain and the curves nearly coincide, with little permanent deformation - direct evidence that the network still recovers under repeated loading.
| Indicator | Value | Reference |
|---|---|---|
| PVA/UPE matrix | almost no shielding effectiveness | - |
| Filler added, not crosslinked | 26.9 dB | - |
| EMI SE after crosslinking | 37.1 dB | commercial level cited in the paper is 20 dB |
| SSE | 1515.53 dB·cm⁻¹ | normalized to density |
| SSE/t | 6314.68 dB·cm²·g⁻¹ | normalized to density and thickness |
| After 100 compressions | 34.7 dB (93.4% retention) | 20%-80% compression cycles |
The test band is X-band, 8.2-12.4 GHz, on samples 5 cm in diameter. The shielding mechanism is assigned to three contributions: the magnetic-dielectric synergistic network, dielectric loss, and multiple reflection and scattering. After crosslinking the continuity of the conductive filler improves visibly, isolated filler is almost absent and most of it is tightly embedded in the film - the direct reason conductivity rises from 4.8 S/cm to 13.7 S/cm (close to three times), and the underlying reason shielding effectiveness takes another step up.
| Treatment | Condition | Result |
|---|---|---|
| Strong acid | hydrochloric acid, pH 1, 30 min | EMI SE falls by less than 3 dB after each treatment and stays above 33 dB |
| Strong alkali | sodium hydroxide, pH 14, 30 min | |
| Deep cold | liquid nitrogen (-196 ℃), 20 min | |
| High temperature | 200 ℃ ambient, 5 min |
After 6 min at 200 ℃ the micrographs show nearly half of the UHMWPE partially melted and collapsed, yet the film keeps its overall shape - the crosslinked network acts as a scaffold here. For a shielding film that has to survive reflow soldering, or serve outdoors or in automotive environments, this figure is more informative than a peak value measured at room temperature.
The conductivity gain from crosslinking adds a thermal management function. The film reaches its saturation temperature within a short time at 2.0 V, 3.0 V and 4.0 V; at 4.0 V the surface saturates at 58.1 ℃. It climbs above 50 ℃ within 100 s of power on and holds stable for 2500 s. Saturation temperature scales linearly with the square of voltage, so a set temperature can be held by adjusting voltage. Photothermally, saturation is reached within 40 s at different light intensities, and the surface temperature barely changes over 200 s of continuous irradiation; at 400 mW/cm² the infrared image shows an even distribution. The paper points to smart wearables and medical hyperthermia as application directions.
State the conclusions in full before use:
| Model | Wavelength | Tubes and chamber | Use |
|---|---|---|---|
| UCL-3500 (the 254 nm version stocked here) | 254 nm (UV-C) | 6 × 15 W; chamber 44×34×15 cm | nucleic acid membrane crosslinking, PCR decontamination, UV mutagenesis and other work judged on UV-C |
| UCL-3500M | 302 nm (UV-B) | 6 × 15 W; same chamber | cases that need longer-wavelength activation while limiting photodamage |
| UCL-3500L | 365 nm (UV-A) | 6 × 15 W; same chamber | photoinitiated crosslinking of the kind described here, photosensitizer activation, photoaffinity labeling; penetrates organic films better |
| UCL-3200 | 254 nm | 6 × 8 W; chamber 32×26×15 cm | single films, samples up to A4, routine throughput |
| UVP CL-3000 | 254 nm (302 and 365 nm versions also available) | 6 × 8 W; door-opening type with viewing window and door lock | sites with an existing UVP/Analytik Jena equipment list that need to stay with the brand |
There are two decisions. Decide the version by wavelength before anything else (254 or 365 nm, depending on initiator absorption and film thickness), then choose the chamber by sample volume per run (UCL-3200 for work up to A4, the 44×34 cm chamber of the UCL-3500 series for several films or large substrates). Energy mode (0.000-9.999 J/cm², nine preset programs) is available on both the UCL-3500 and the UCL-3200, and is the function to keep for work specified by dose.
No - it is a matter of version. UCL-3500 covers three wavelength versions of the same instrument: the 254 nm UCL-3500, the 302 nm UCL-3500M and the 365 nm UCL-3500L. The paper states 365 nm explicitly, so the corresponding unit is the UCL-3500L. Write both the model and the wavelength into the technical requirement when purchasing, which avoids a default shipment of the 254 nm version.
Start by separating three quantities: the power setting of the instrument, the measured irradiance inside the chamber, and the cumulative energy. The 4.68×105 J/m² reported in the paper is four times the value calculated from "6.5 mW/cm² × 30 min", which shows that the power setting on paper is not the cumulative dose the film actually receives. Reproduce from the energy reading; if just an intensity reading is available, measure it with a radiometer in the plane where the sample sits, convert with dose = intensity × time, and allow for the distribution differences inside the chamber.
It can, with one extra calibration step: measure the actual intensity at the sample plane with a UV radiometer, calculate the time for the target dose, and record both the tube hours and the measured intensity in the notebook. After the tubes age, the same number of seconds delivers a lower dose, and the conditions drift quietly - the usual trap in time mode.
The paper irradiates directly after soaking and dries afterwards. Irradiating wet (with ethanol present) has the advantage that oxygen enters the film less readily, so oxygen inhibition is weaker and radicals live longer; with dry irradiation the surface is inhibited by oxygen before it can react, efficiency falls, and a higher dose is usually needed. Either route is workable, but fix one and write it into the method, otherwise batches are not comparable.
Usually the dose or the crosslinker concentration is too high. Bring the cumulative energy down one step before anything else (from 46.8 J/cm² toward 20 J/cm², for instance), then consider reducing TMPTMA from 1.0% to 0.5%. Drying too fast also worsens curling; slow drying in a fume hood, or drying under a weight, helps. The paper's own result - elongation lower than the filler-free base film but still close to it - indicates that such a balance point exists.
The UCL-3500 series has a 44×34×15 cm chamber and several films can be laid flat without difficulty, but intensity differs between positions near the chamber wall and near the center of the tubes. Map the chamber distribution once with a radiometer, place the critical samples in the center zone, and add time separately for edge samples according to the measured intensity. Record the dose for each position in a batch rather than sharing one set value.
All process and performance data in this case are taken from the paper cited above; equipment model and irradiation parameters follow its methods section. The notes on wavelength version correspondence and on the difference in cumulative energy conventions are our own reading added during preparation, and do not represent the views of the original authors.
This article belongs to the research laboratory techniques section.
Other sections on the same product line include UV crosslinkers (dose set in J/cm², the number that matters for both nucleic acid crosslinking and materials curing) and life science instruments (the entry point for fluorescence imaging, UV crosslinking, UV transmission and cell irradiation).
Models mentioned in this article: UCL-3500, UCL-3200, UVP CL-1000. Ranges, accuracy and accessory lists for each model are given on the product pages.
Email: info@xinysun.com
Tel: 131-2093-6518
Address: 1525 Minqiang Road, Bldg 26, Songjiang, Shanghai
Copyright 2026 Xinysun Instrument (Shanghai) Co., Ltd. All rights reserved.
ICP:ICP-2026046447
